How Do Social Interaction Networks Influence Peer Impressions Formation? A Case Study

Due to their lack of physical interaction, Free and Open Source Software (FOSS) participants form impressions of their teammates largely based on sociotechnical mechanisms including: code commits, code reviews, mailing-lists, and bug comments. These mecha

  • PDF / 536,929 Bytes
  • 10 Pages / 439.363 x 666.131 pts Page_size
  • 79 Downloads / 194 Views

DOWNLOAD

REPORT


Abstract. Due to their lack of physical interaction, Free and Open Source Software (FOSS) participants form impressions of their teammates largely based on sociotechnical mechanisms including: code commits, code reviews, mailing-lists, and bug comments. These mechanisms may have different effects on peer impression formation. This paper describes a social network analysis of the WikiMedia project to determine which type of interaction has the most favorable characteristics for impressions formation. The results suggest that due to lower centralization, high interactivity, and high degree of interactions between participants, the code review interactions have the most favorable characteristics to support impression formation among FOSS participants. Keywords: Open Source, OSS, FOSS, social network analysis, collaboration.

1

Introduction

Impression Formation (i.e., obtaining an accurate perception of teammates’ abilities) is difficult for Free Open Source Software (FOSS) developers because of the lack of physical interaction with their distributed, virtual teammates. Inaccurate perceptions of teammates’ abilities and expertise may reduce the team’s productivity because some developers improperly disregard the opinions or contributions of certain teammates. Due to the lack of physical interaction, members of distributed teams form impressions based on tasks [16]. This task-focus increases the potential effects of socio-technical interactions (i.e. social interactions facilitated by technological solutions) on impression formation. Common FOSS socio-technical interaction mechanisms include: project mailing-lists, bug repository, code review repository, and code repository. Each of these socio-technical mechanisms facilitates different types of interactions among project participants. The results of our recent large-scale survey of FOSS developers suggest that developers believe code review is beneficial in the impression formation process [6]. Our hypothesis in this work is that due to the nature of interaction supported by each of these mediums, some may be more beneficial than others for impression formation. We conducted a social network analysis (SNA) of the interactions facilitated by each socio-technical mechanism to determine which of those mechanisms tended to support impression formation. L. Corral et al. (Eds.): OSS 2014, IFIP AICT 427, pp. 31–40, 2014. c IFIP International Federation for Information Processing 2014 

32

A. Bosu and J.C. Carver

A social network is a theoretical construct that represents the relationships between individuals, organizations, or societies. These networks are typically modeled as a graph in which vertices represent individuals or organizations and edges represent a relationship between individuals or organizations. SNA focuses on understanding the patterns of interactions and the relative positions of individuals in a social settings [10]. SNA is commonly used in many domains, e.g., sociology, biology, anthropology, communication studies, information science, organizational studies,